DESCRIPTION (provided by investigator): Colorectal cancer (CRC) is the second leading cause of cancer death in the US. Approximately 20-35 percent of CRC is attributable to inherited factors, and identification of associated genetic variants is important to elucidate mechanisms underlying this disease. First results from genome-wide association studies (GWAS) have demonstrated considerable success in identifying genetic variants associated with various common complex diseases, including CRC. However, the full complement of genetic susceptibility loci and the roll of gene- environment interactions underlying CRC are still largely unexplained. Large sample sizes, both for the initial genome-wide scan, as well as for the replication in independent study populations, have become crucial in identifying and establishing true associations. Further, identification and characterization of genetic risk variants will be enhanced if key environmental risk factors are also taken into account. To accelerate the discovery of CRC- related variants, we are proposing a genome-wide scan of population-based association studies of CRC (G- PAC). The goal of this project is to expand the scope of our current GWAS to encompass an additional 3,811 CRC cases and 3,811 controls to be genotyped on the Illumina HumanHap 610k genome-wide scan platform. We will pool GWAS data from samples with our existing GWAS data, bringing our total to 5,645 cases and 11,412 controls from well-described population based cohorts and a case-control study, including the Women's Health Initiative, Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial, Nurses'Health Study, Health Professional Follow-up Study, Physician's Health Study, Vitamin and Lifestyle Study, and a population-based case-control study. We will then use this data in two ways to identify potential CRC susceptibility genes: a) We will perform a meta-analysis combining our G-PAC data with other GWAS carried out by other groups;b) we will do a genome-wide search for genetic variants that are modified by well known risk factors for CRC, such as obesity, physical activity, non-steroidal anti-inflammatory drugs, hormone use, folate, calcium, red and processed meat, alcohol, or smoking. This two prong analysis will provide a powerful means to select the most promising genetic variants, which will be followed up in an independent replication study. For replication we will genotype 7,600 variants in 3,526 cases and 3,618 controls from two population-based case-control studies. The large sample size and detailed exposure and outcome ascertainment in G-PAC study populations provide a unique resource to identify several new CRC susceptibility genes/loci and to elucidate gene-environment interactions. We expect that our findings will enhance our understanding of the genetic susceptibility and molecular mechanisms of colorectal carcinogenesis, thus leading to improved preventive strategies. This applications is in line with the goals of the Recovery Act, accelerating the tempo of science and leading to significant job retention and creation (equivalent to18.8 jobs). PUBLIC HEALTH RELEVANCE: Colorectal cancer is the second leading cause of cancer death in the United States. This study will identify genetic variants associated with this disease, and will examine whether environmental factors, including smoking, medications, alcohol, physical activity, or diet change the risk of colorectal cancer related to these genetic variants. We expect our findings to enhance our understanding of the mechanisms underlying colorectal carcinogenesis. In turn, this will lead to improved prevention and treatment strategies.